A Digital Twin is a virtual copy or representation of a physical object. Given the same input, both should produce the same output. The data flow between them should keep both objects in the same state by automatically reflecting changes from one of the objects to the other. Despite the popularity of the topics, there is a lack of standardization and no widespread adoption of techniques or architectures, resulting in ad-hoc implementations for each use case. In this work, we propose a methodology for the generation of digital twins based on the definition of their life cycle, with phases that include training, operation and retraining. To allow efficient management of resources, each phase of the Digital Twins will be implemented as a PyCOMPSs workflow. The goal of proposing methodologies and architectures is to simplify and make the design and development of digital twins more robust; it also makes it possible to extend existing ones, replace their components with more advanced ones, and support their reuse. Additionally, workflow deployment is usually a difficult task that requires specialized staff and time, so we propose using a specific tool for the deployment of the different workflows and their software requirements.
Fernando Vázquez Novoa (Tue,) studied this question.